5 ms·

In my mind STEM topics exist on a scale that has math at one end, then there's a slight pause/discontinuity (as in point 2 in the underpants gnome business plan

by bborud 5y ago

In my mind STEM topics exist on a scale that has math at one end, then there's a slight pause/discontinuity (as in point 2 in the underpants gnome business plan) before it continues with physics, chemistry, biology and onwards to god knows what. As you move along this axis stuff feels more and more complex, messy and hard to nail down definitively. More practical. More prone to get tripped up by silly mistakes and things you never imagined you had to think about. The new compounds they discovered in the old samples kind of illustrates how much hard work is required to get good answers. You can't just suck on your pipe, scratch your quill on some parchment and exert yourself mentally, you have to scrape gunk out of test-tubes and be really careful and thorough in a highly practical sense. The world is imprecise and gnarly and you kind of have to hope for the best. I really admire people who are capable of wringing results out of goo and specks of dust using gadgets that require calibration that has to be able to distinguish actual good signal from residues of the danish you had for breakfast. (Example not entirely random since one of our engineers eating a danish for lunch in his office made the electrochemical sensors he was working on go absolutely apeshit. Kind of good to know before you evacuate a whole industrial site and send in the people in yellow hazmat suits)

wodenokoto 5y ago
> In my mind STEM topics exist on a scale that has math at one end, then there's a slight pause/discontinuity (as in point 2 in the underpants gnome business plan) before it continues with physics, chemistry, biology and onwards to god knows what. Man there really is an XKCD for everything: https://xkcd.com/435/ https://xkcd.com/435/
- ARandomerDude 5y ago
  "Now one speculative [i.e., theoretical] science is said to be nobler than another, either by reason of its greater certitude, or by reason of the higher worth of its subject-matter...Of the practical sciences, that one is nobler which is ordained to a further purpose, as political science is nobler than military science; for the good of the army is directed to the good of the State." Thomas Aquinas, ST I, q. 5., a. 1
8note 5y ago
I think philosophy sits on either end, making it a circle
snowwrestler 5y ago
I try to keep this in mind when reviewing predictions, speculation, and fiction about artificial intelligences that get super intelligent. There’s more to knowledge than just calculating faster. Hypotheses need to be tested by experimentation; theories must be supported by evidence. (This is not my original idea, I read it in an essay from an AI researcher.) This is also why there is more to science than statistical analysis. Stats help us understand evidence we have collected. But at some point a new idea needs a new test in order to know if it is useful or correct.
shmageggy 5y ago
> In my mind STEM topics exist on a scale that has math at one end, then there's a slight pause/discontinuity (as in point 2 in the underpants gnome business plan) before it continues with physics, chemistry, biology and onwards to god knows what. https://xkcd.com/435/ https://xkcd.com/435/
- HPsquared 5y ago
  And mathematics is just applied philosophy, philosophy is just applied language, and language is again one of those "social studies"...
- bborud 5y ago
  Holy crap! :)
  - LanceH 5y ago
    And the joke is way, way older than that.
    - bborud 5y ago
      Whether or not it is a joke probably depends on how you label the axis along which you distribute these topics. :) Observe that as you move from one side to the other (math to biology and onwards), you are essentially dragging along a good portion of the stuff as prerequisites you have to know a bit about. By the time you reach something like immunobiology you can pretty much build a house out of the textbooks and go live in it. :)
      - LanceH 5y ago
        The joke as I heard it in the 80's -- and I have no reason to think it was new then -- went something like: Physics is just applied math, chemistry is applied physics, etc... I've always liked the view that Physics is just experimental discovery of the axioms "chosen" for our universe. Physics without experimentation just being a specific subset of math.
azalemeth 5y ago
I've been lucky enough to work with a few highly skilled organic chemists, and I know enough chemistry to know that I could never, ever join them as such. It's such a niche, specialised, and important and yet under-recognised set of skills. A few examples. The order in which you add nominally unimportant reagents to a reaction can significantly matter – the enthalpy of mixing is non-zero, and so if you dissolve A and then B in solvent C before letting it reflux for five hours (say), it will be a different initial condition to adding B, then A. Many (most?) chemists weigh vials before adding a mass of x g/mg of a compound, in the process of adding that, and then again when they have removed "x" grams. Lids are pared with volumetric flasks. Tiny variations matter. Room humidity, temperature, and the history of the use of the equipment involved may change the answers. I had one graduate student (who came top in her year in chemistry) make an isotopically labelled molecule for me, with a ~70% yield and a ~100% labelling purity. A few years later, a biochemist following the "recipe" she developed (and beautifully illustrated!) was unable to get above 10% yield and about 80% purity. I never really was able to work out why. Any "debugging" conversation starts with "tell me exactly -- what did you do?". Things occasionally don't go "as well" when certain contaminants are (not) present. The debugger for reality sucks. And still, despite all the fiddly nature of the work, it's downright dangerous. Organic chemists, as a species, live less long. They get blown up and gassed. I've been in a building when a "small scale" reaction has exploded and you feel the vibrations of it – nobody was seriously injured because the fume hood's shock sensors detected the azide derivatives decomposing rapidly and, within microseconds, dumped a load of cold CO2 down the front of the hood, stopping combustion and redirecting/reducing the blast. Another acquaintance of mine managed to have the building evacuated after ordering a 10 ml bottle of a compound he distilled from a hops extract and identified via GC-MSMS and MS^n spectrometry as part of a project trying to understand beer foam (and make better non-alcoholic beers; this is in about 2000 or 1999, before "good" alcoholfrei bier was available) – he naïvely opened the lid to see what it smelt like outside of the fume hood, and immediately uncontrollably vomited. So did his neighbours. Then the rest of the room. Then the floor. They closed it over a weekend, opened all the doors and windows, put in industrial fans, and it still stank for months. Chemists do important, undervalued work. They make drugs, materials, batteries. They do everything from quantum mechanics to what I still consider to be not far removed from alchemy. Some of them are slightly odd people, but they are usually very fun to share a drink with...

by bborud 5y ago

wodenokoto 5y ago

> In my mind STEM topics exist on a scale that has math at one end, then there's a slight pause/discontinuity (as in point 2 in the underpants gnome business plan) before it continues with physics, chemistry, biology and onwards to god knows what. Man there really is an XKCD for everything: https://xkcd.com/435/ https://xkcd.com/435/

ARandomerDude 5y ago
"Now one speculative [i.e., theoretical] science is said to be nobler than another, either by reason of its greater certitude, or by reason of the higher worth of its subject-matter...Of the practical sciences, that one is nobler which is ordained to a further purpose, as political science is nobler than military science; for the good of the army is directed to the good of the State." Thomas Aquinas, ST I, q. 5., a. 1

8note 5y ago

I think philosophy sits on either end, making it a circle

snowwrestler 5y ago

I try to keep this in mind when reviewing predictions, speculation, and fiction about artificial intelligences that get super intelligent. There’s more to knowledge than just calculating faster. Hypotheses need to be tested by experimentation; theories must be supported by evidence. (This is not my original idea, I read it in an essay from an AI researcher.) This is also why there is more to science than statistical analysis. Stats help us understand evidence we have collected. But at some point a new idea needs a new test in order to know if it is useful or correct.

shmageggy 5y ago

HPsquared 5y ago
And mathematics is just applied philosophy, philosophy is just applied language, and language is again one of those "social studies"...
bborud 5y ago
Holy crap! :)
- LanceH 5y ago
  And the joke is way, way older than that.
  - bborud 5y ago
    Whether or not it is a joke probably depends on how you label the axis along which you distribute these topics. :) Observe that as you move from one side to the other (math to biology and onwards), you are essentially dragging along a good portion of the stuff as prerequisites you have to know a bit about. By the time you reach something like immunobiology you can pretty much build a house out of the textbooks and go live in it. :)
    - LanceH 5y ago
      The joke as I heard it in the 80's -- and I have no reason to think it was new then -- went something like: Physics is just applied math, chemistry is applied physics, etc... I've always liked the view that Physics is just experimental discovery of the axioms "chosen" for our universe. Physics without experimentation just being a specific subset of math.

azalemeth 5y ago

I've been lucky enough to work with a few highly skilled organic chemists, and I know enough chemistry to know that I could never, ever join them as such. It's such a niche, specialised, and important and yet under-recognised set of skills. A few examples. The order in which you add nominally unimportant reagents to a reaction can significantly matter – the enthalpy of mixing is non-zero, and so if you dissolve A and then B in solvent C before letting it reflux for five hours (say), it will be a different initial condition to adding B, then A. Many (most?) chemists weigh vials before adding a mass of x g/mg of a compound, in the process of adding that, and then again when they have removed "x" grams. Lids are pared with volumetric flasks. Tiny variations matter. Room humidity, temperature, and the history of the use of the equipment involved may change the answers. I had one graduate student (who came top in her year in chemistry) make an isotopically labelled molecule for me, with a ~70% yield and a ~100% labelling purity. A few years later, a biochemist following the "recipe" she developed (and beautifully illustrated!) was unable to get above 10% yield and about 80% purity. I never really was able to work out why. Any "debugging" conversation starts with "tell me exactly -- what did you do?". Things occasionally don't go "as well" when certain contaminants are (not) present. The debugger for reality sucks. And still, despite all the fiddly nature of the work, it's downright dangerous. Organic chemists, as a species, live less long. They get blown up and gassed. I've been in a building when a "small scale" reaction has exploded and you feel the vibrations of it – nobody was seriously injured because the fume hood's shock sensors detected the azide derivatives decomposing rapidly and, within microseconds, dumped a load of cold CO2 down the front of the hood, stopping combustion and redirecting/reducing the blast. Another acquaintance of mine managed to have the building evacuated after ordering a 10 ml bottle of a compound he distilled from a hops extract and identified via GC-MSMS and MS^n spectrometry as part of a project trying to understand beer foam (and make better non-alcoholic beers; this is in about 2000 or 1999, before "good" alcoholfrei bier was available) – he naïvely opened the lid to see what it smelt like outside of the fume hood, and immediately uncontrollably vomited. So did his neighbours. Then the rest of the room. Then the floor. They closed it over a weekend, opened all the doors and windows, put in industrial fans, and it still stank for months. Chemists do important, undervalued work. They make drugs, materials, batteries. They do everything from quantum mechanics to what I still consider to be not far removed from alchemy. Some of them are slightly odd people, but they are usually very fun to share a drink with...